Comment on Brad DeLong’s GDP Estimate under Obama vs. McCain
August 23, 2008
Misleading Claim from Furman and Goolsbee
In last week's now famous Wall Street Journal op-ed by Obama advisors Austan Goolsbee and Jason Furman, they make the following claim in support of their candidate:
In contrast, Sen. McCain has put forward the most fiscally reckless presidential platform in modern memory. The likely results of his Bush-plus policies are clear. As Berkeley economist Brad Delong has estimated, the McCain plan, as compared to the Obama plan, would lower annual incomes by $300 billion or more in real terms by 2017, costing the typical worker $1,800 or more due to the effect of large deficits on national savings and thus capital formation. Sen. McCain's neglect of critical public investments would further impede economic growth for decades to come.
They are referencing this Brad DeLong back-of-the-envelope calculation of what he projects would be the debt-imposed effect on long-run GDP from likely Obama and McCain fiscal policies. But the fact that Goolsbee and Furman use the phrases "McCain plan" and "Obama plan" to describe DeLong's estimate is factually incorrect.
The fact of the matter is that DeLong did not take Obama's or McCain's fiscal plans as presented by the campaigns or estimated by Tax Policy Center to determine his estimates. DeLong made the following much simpler assumptions (possibly out of necessity) that essentially had little to do with the specific plans being offered by the two campaigns:
McCain: Begin by assuming that John McCain's fiscal policy is likely to be standard Republican fiscal policy. Since 1981, standard Republican fiscal policy has been to increase the ratio of gross federal debt to GDP by nearly 2% per year.
Obama: And suppose, last of all that the alternative to a McCain presidency-a Barack Obama presidency-will not be able to match the Democratic post-WWII average in fiscal-policy prudence but will instead simply hold the gross federal debt-to-GDP ratio constant.
So the relevant question becomes "Is the expected value of McCain's fiscal policies to increase the debt-to-GDP ratio equal to 2 percent each year?" Who knows? That depends on a host of factors. McCain's fiscal policy record (as a senator) is pretty good by deficit-hawk standards. He opposed the Medicare Part D expansion, and he continuously opposed farm subsidies, two policies that are far more than mere pork-barrel spending (which he too often claims as being a 'cure all'). On the other hand, it's worth pointing out that McCain's foreign policy views could lead to an expanded defense budget.
Comment on DeLong's methods
Now I personally believe that this issue of the long-run impact of the national debt is an important one that has been overlooked for too long by those who believe that "starve the beast" works well to promote national welfare (see latest work of two other Berkeley economists, Romer and Romer on this topic). As DeLong points out in this piece: "…a growing debt-to-GDP ratio crowds out investment, as resources that would otherwise go to fund productive investment instead support private or public consumption."
It should be noted that DeLong is a leading scholar on this topic, although he does support Obama for president. (For those interested in learning macroeconomics with a special emphasis on capital, DeLong's undergraduate textbook is a terrific introduction to the debt-GDP issue along with the related issues of international capital flows and long-run growth) But one missing piece in DeLong's paper is that he fails to mention the important fact that not all deficits have the same adverse effect on long-run GDP.
While Tax Policy Center estimates that McCain would increase the debt by a rather significant amount (more than Obama), much of that static estimate is due to lower taxes on capital. If McCain's lower taxes on capital would increase savings today, then part of the harm done to GDP by McCain's deficits would be offset. In other words, the national savings rate would not fall by the full amount of the debt increase. Despite what DeLong says above, these tax cuts that caused the higher debt wouldn't necessarily be going to finance current consumption.
On the other hand, Obama's tax plan is essentially redistribution from high-income earners to a group that has a very high marginal propensity to consume. So even under a static score, while Obama's fiscal policies may be less costly than McCain's, Obama's would have little additional effect of increasing national savings.
Overall, I think it's safe to say that $1 added to the debt by McCain's likely fiscal plan would have a smaller adverse effect on GDP than $1 added to the debt under Obama's fiscal plan.
(Now this gets into the secondary question of whether GDP is the relevant measure of economic well-being, which it is far from. On this, I'll point out two defenses for the Obama team: (1) GDP aggregates don't account for the distribution of that GDP (national income), which Obama has made a priority and which society seems to be concerned with. (2) Obama's tax hikes on labor may lower work effort, but GDP doesn't measure the value of likely substitutes to that work, which are domestic production and leisure, the value of which are not zero. Accounting for such items that are included in social welfare is difficult.)